Decomposition: What It Tests and a Repeatable Framework
Apply a five-pass decomposition framework to turn broad goals into actionable decisions. Identify the user and decision, define goals and constraints, gather relevant evidence, surface consequential unknowns, and create a bounded restatement of the problem. Ask critical questions that could change the scope of the work, ground decisions in evidence and clear objectives, and iteratively refine the problem as new information emerges.
We'll cover the following...
Catalog species in an unfamiliar environment” has at least two reasonable interpretations:
Help a researcher identify specimens.
Help a field coordinator choose tomorrow’s survey sites.
Both could justify a catalog, but they need different evidence and different measures of success.
Decomposition turns a broad goal into a bounded decision that an authorized person can make with the evidence available. The first useful output is a defensible problem statement. Choosing a database or a classifier comes after that boundary is clear.
What the round tests
Public candidate reports describe the decomposition round as an open-ended conversation, usually with little or no code. The prompt is deliberately vague. A strong candidate doesn’t jump to a solution. They establish whose problem is worth solving, what action is possible, and whether the evidence supports it.
In practice, that means showing four things:
You ask questions whose answers change the work.
You separate what is known from what is assumed.
You notice when data looks complete but can’t support the decision.
You end with a scoped statement someone could act on.
A repeatable framework: Five passes
Each pass produces something you can say aloud in one or two sentences:
...